ANALISIS E INTERPRETACION DE RESULTADOS

6.2.1 DZ may be homologous to caudal auditory fields in the primate

Although parcellation of auditory cortical areas into core versus belt regions has been proposed and generally accepted in the primate literature (Hackett et al., 2001), no such formal separation has been proposed for auditory cortical regions in the cat (Read et al., 2001). It has been suggested that DZ forms part of a functional belt region of auditory cortex because of the complexity and longer latency of responses in DZ in comparison to core fields A1 and AAF, as well as evidence that DZ plays a role in temporal (He et al., 1997) and spatial auditory processing (Stecker et al., 2005; Malhotra et al., 2008). However, no homologous structure in primate auditory cortex has previously been proposed for DZ. In the past, homology between cat and macaque visual systems has been suggested using a set of criteria including behaviorally determined function, cortical position, electrophysiological responses, and anatomical cortical and thalamic connectivity (Payne, 1993).

On the basis of these criteria, many parallels can be drawn between DZ in the cat and caudal regions of primate auditory cortex. Behaviorally, “what” and “where” auditory processing streams have previously been identified for both the cat (Lomber and Malhotra, 2008) and the primate (Romanski et al., 1999; Rauschecker and Tian, 2000). In the primate, the caudolateral (CL) and

178

caudomedial (CM) areas are known to be involved in processing auditory spatial information (Rauschecker, 1998; Rauschecker and Tian, 2000; Recanzone, 2000a; Woods et al., 2006; Kusmierek and Rauschecker, 2014), as is DZ (Stecker et al., 2005; Malhotra et al., 2008). In terms of cortical position, caudal fields CM and CL in the monkey are located dorso-posteriorly to core fields A1 and the rostral field (R). DZ is also located dorso-posteriorly to A1 and AAF in the cat, and AAF of the cat is considered to be homologous to primate field R (Rauschecker et al., 1997).

Electrophysiologically, response properties in caudal fields of the primate share a number of similarlities with DZ. Receptive fields in CM are more broadly tuned than the sharp frequency tuning found in core areas A1 and R; (Merzenich and Brugge, 1973; Morel et al., 1993; Kosaki et al., 1997)), with neurons in CM responding better to noise than to tone stimuli (Recanzone, 2000b), and neurons in CL responding better to band-passed noise than to pure tones (Rauschecker et al., 1995). Monotonic rate-level functions were also identified in CM of primate auditory cortex, with the vast majority characterized as monotonic or saturating (Kajikawa et al., 2005). Additionally, Woods et al. (2006) have shown that both CM and CL neurons respond most strongly at the highest sound intensity level presented. Similarly, Chapter 3 showed that DZ neurons respond better to noise than to tones, and also showed a dominance of monotonic rate-level functions. Perhaps most convincing is the effect of A1 ablation on neuronal responses in R and CM (Rauschecker et al., 1997). A1 ablation had no effect on neuronal responses in R, but abolished pure tone responses in CM, while responses to more complex stimuli were preserved, albeit weaker in magnitude. These results correspond well with those of Chapter 3 in which A1 deactivation strongly reduces receptive field bandwidths and increases neuronal thresholds in DZ, but has no effect on peak responses in AAF (Carrasco and Lomber, 2010). Finally, visual modulation of auditory responses in CM and CL have been shown using functional imaging (Kayser et al., 2007) and electrophysiological recordings (Kayser et al., 2008). While auditory-somatosensory integration has not been

179

demonstrated in either of these fields, somatosensory inputs capable of eliciting responses to median nerve stimulation have been documented in this area (Schroeder and Foxe, 2002). Again, these results correspond well with the visual and somatosensory modulation of auditory responses in DZ in Chapter 4.

Finally, both DZ (Lee and Winer, 2008a) and CM/CL (Rauschecker et al., 1997; de la Mothe et al., 2012) receive input from mainly dorsal regions of auditory thalamus. These data stand in stark contrast to the strong projections from ventral regions of auditory thalamus to A1 and AAF in the cat (Lee and Winer, 2008a) and A1 and R in the primate (Molinari et al., 1995; Rauschecker et al., 1997). Furthermore, A1 shares dense, reciprocal connectivity with fields CM and CL (Kaas and Hackett, 2000), as does DZ (Lee and Winer, 2008a).

Despite these similarities, a few discrepancies between caudal areas in the primate and DZ in the cat should be noted. Response latencies of neurons in CM and CL, for example, have been shown to be shorter or similar to those of A1 (Camalier et al., 2012; Kusmierek and Rauschecker, 2014), whereas in the cat, only AAF neurons have shorter response latencies than neurons in A1 (Carrasco and Lomber, 2009). CM is also roughly tonotopically organized in the primate (Morel et al., 1993), whereas DZ is traditionally considered to be non-tonotopic (Lee and Winer, 2011). While it could be argued that PAF and fAES may also be considered as candidates for homology with CM based on the behavioral similarity of their spatial processing roles (Malhotra et al., 2007), fAES and PAF both violate some of the above-metioned criteria for the establishment of homology. fAES violates the cortical position criterion, being located antero- ventrally to A1 and AAF, as well as the anatomical cortical connectivity criterion, as it does not receive a strong projection from A1 (Lee and Winer, 2008a). PAF violates thalamic connectivity criterion, as the strongest auditory thalamic projection it receives is from the ventral division of MGN, and as such, has been considered homologous with core regions of primate auditory cortex (Hackett et

180

al., 2011). Furthermore, neither fAES or PAF neurons have been shown to be modulated by either visual or somatosensory stimuli.

Therefore, the work presented in this thesis adds to a body of literature documenting shared behavioral, topographic, electrophysiological and anatomical similarities with caudal fields CM/CL of the primate, and as such, fulfil the criteria outlined above for possible homology with these fields.

6.2.2 DZ is the most extensively-documented model of cross-modal plasticity in mammalian cortex to date

In addition to the previously published behavioral evidence of cross-modal reorganization in DZ (Lomber et al., 2010), this thesis adds a number of novel insights which together make this region the most comprehensive model of cross-modal plasticity in mammalian cortex. This thesis documents anatomical changes in connectivity in deaf animals (Chapter 2) which corroborate previous behavioral findings (Lomber et al., 2010), electrophysiological evidence of cross- modal plasticity (Chapter 5), and evidence of multisensory processing in DZ at multiple scales of neuronal activity, which allow for unique insights into the cross- modal reorganization that takes place following deafness (Chapter 4).

While some neuroimaging studies have documented behavioral evidence of superior performance in blind or deaf individuals, and have located brain regions involved in mediating these enhanced abilities (e.g. Sadato et al., 1996), changes in the functional connectivity of the cerebral cortex following blindness or deafness have not been documented, and as such, the mechanisms that give rise to this plasticity remain murky. With respect to V1, some anatomical changes have been documented in the opossum (Karlen et al., 2006); however it remains unknown whether these changes are generalizable to species phylogenetically more closely related to humans.

Similarly, in the animal literature, while some studies have demonstrated a behavioral improvement in the performance of a task in sensory-deprived

181

animals and have correlated it with electrophysiological evidence of cross-modal plasticity (e.g. Izraeli et al. 2002), only Lomber and colleagues (2010) have conclusively localized enhanced ability to a particular brain region. Aside from this, although a few studies in cat cortex have correlated behavioral enhancements with electrophysiological findings (Korte and Rauschecker, 1993; Rauschecker and Korte, 1993; Rauschecker and Kniepert, 1994, Meredith et al., 2011), again, the connectional changes that give rise to this plasticity remain undetermined. Furthermore, the multisensory processing capabilities of these areas remain largely uninvestigated and could provide important information regarding the influence of other sensory modalities on the area of interest in non- deprived animals, particularly in the absence of documented connectional changes following deprivation.

Together, this thesis provides original insights into the structure and function of DZ in both hearing and deafened animals. These findings conclusively demonstrate that DZ receives active visual inputs in hearing animals that are strengthened following deafness to contribute to the reorganization of DZ as a region involved in visual motion processing. This research has spawned a number of interesting avenues for future experimentation to further elucidate our understanding of the principles underlying basic sensation as well as cross- modal plasticity.